Comparison of specific segmentation methods used for copy move detection

<p><span lang="EN-US">In this digital age, the widespread use of images and availability image editors have made credibility controversial. To confirm many forgery detection types are arises, copy-move is consisting transforming any by duplicating a part image, to add or hide existing objects. Several methods been proposed in literature detect forgery, these key point-based block-based find duplicated areas. However, drawback ability handle smooth region. In addition, segmentation plays vital role changing representation meaningful form for analysis. Hence, we execute comparison study based on two clustering algorithms (i.e., k-means super pixel with density-based spatial applications noise (DBSCAN)), paper compares term accuracy detecting regions images. K-means shows better performance compared DBSCAN other techniques literature.</span></p>